MCPcopy Create free account

hub / github.com/UbiquitousLearning/mllm / functions

Functions5,971 in github.com/UbiquitousLearning/mllm

⨍ Functions 5,971◇ Types & classes 1,855↳ Endpoints 4
↓ 2 callersFunctionhasAVX512CD
mllm/utils/CPUArchHelper.hpp:285
↓ 2 callersFunctionhasAVX512DQ
mllm/utils/CPUArchHelper.hpp:291
↓ 2 callersFunctionhasAVX512F
mllm/utils/CPUArchHelper.hpp:273
↓ 2 callersFunctionhasAVX512VL
mllm/utils/CPUArchHelper.hpp:297
↓ 2 callersMethodhasDispatcher
mllm/engine/DispatcherManager.cpp:33
↓ 2 callersFunctionhasNEON
mllm/utils/CPUArchHelper.hpp:211
↓ 2 callersFunctionhasSSE
mllm/utils/CPUArchHelper.hpp:224
↓ 2 callersFunctionhasSSE2
mllm/utils/CPUArchHelper.hpp:230
↓ 2 callersFunctionhasSSE3
mllm/utils/CPUArchHelper.hpp:236
↓ 2 callersFunctionhasSSE4_1
mllm/utils/CPUArchHelper.hpp:248
↓ 2 callersFunctionhasSSE4_2
mllm/utils/CPUArchHelper.hpp:254
↓ 2 callersFunctionhasSSSE3
mllm/utils/CPUArchHelper.hpp:242
↓ 2 callersFunctionhertz_to_mel_slaney
Convert frequency from hertz to mel scale (Slaney implementation)
mllm/models/minicpm_o2_6/audio_preprocessor_minicpmo.hpp:18
↓ 2 callersMethodinfer
(self, prompt: str)
pymllm/backends/qualcomm/transformers/qwen3/runner.py:237
↓ 2 callersMethodinitFromQnnTensor
mllm/backends/qnn/QNNUtils.cpp:529
↓ 2 callersMethodinit_io
mllm/backends/qnn/aot_rt/TokenGenerator.cpp:19
↓ 2 callersMethodinit_tts_module
mllm/models/minicpm_o2_6/modeling_minicpmo.hpp:157
↓ 2 callersFunctioninitializeTemplates
Initialize templates
mllm/models/deepseek_ocr/conversation_preprocess.hpp:259
↓ 2 callersFunctioninsertSession
mllm/c_api/Runtime.cpp:118
↓ 2 callersFunctioninsert_session
(id: str, s: Session)
pymllm/service/rr_process.py:30
↓ 2 callersFunctioniq2_find_best_neighbour
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q2.cpp:297
↓ 2 callersMethodisAlloc
mllm/backends/qnn/QNNUtils.hpp:210
↓ 2 callersFunctionisAndroid
mllm/utils/PlatformRTHelper.hpp:59
↓ 2 callersFunctionisFloat
mllm/backends/qnn/aot/visitor/CastType.cpp:15
↓ 2 callersMethodisMatch
===----------------------------------------------------------------------===// Sigmoid Pattern ===----------------------------------------------------
mllm/backends/qnn/aot/passes/LLMQuantRecipePass.cpp:259
↓ 2 callersFunctionisQnnAvailable
mllm/mllm.cpp:99
↓ 2 callersMethodisRequired
mllm/utils/Argparse.hpp:105
↓ 2 callersMethodisSupportedFP16
mllm/backends/opencl/runtime/OpenCLRuntime.cpp:82
↓ 2 callersMethodis_alive
Checks if the shell process is still running.
pymllm/utils/adb.py:137
↓ 2 callersFunctionis_qnn_aot_on_x86_enabled
()
pymllm/ffi/__init__.py:21
↓ 2 callersMethodlanes
mllm/backends/cpu/kernels/arm/primitives.hpp:68
↓ 2 callersMethodline
mllm-kernel/include/mllm_kernel/source_location.hpp:30
↓ 2 callersFunctionloadExtensionOpset
mllm/mllm.cpp:116
↓ 2 callersFunctionloadImages
* Loads images from conversation messages * * @param conversations JSON array of conversation messages * An example is: * [ *
mllm/models/deepseek_ocr/conversation_preprocess.hpp:344
↓ 2 callersFunctionloadOpPackage
mllm/mllm.cpp:114
↓ 2 callersFunctionmadd
mllm/backends/cpu/kernels/common/llamafile/llamafile_sgemm.cpp:115
↓ 2 callersFunctionmakeLLaMARotaryPosEmbedding
mllm/models/minicpm_o2_6/modeling_chattts.hpp:24
↓ 2 callersFunctionmakeMultimodalPositionEmbedding
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:182
↓ 2 callersFunctionmakeMultimodalPositionEmbedding
mllm/models/minicpm_o2_6/modeling_qwen2vl_for_minicpmo.hpp:181
↓ 2 callersFunctionmakeRotaryPosEmbedding
mllm/models/minicpm4/modeling_minicpm4.hpp:37
↓ 2 callersFunctionmakeVisualRoPEInvFreq
mllm/models/qwen2vl/modeling_qwen2vl.hpp:17
↓ 2 callersFunctionmakeVisualRoPEInvFreq
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:18
↓ 2 callersFunctionmakeVisualRotaryPosEmb
mllm/models/qwen2vl/modeling_qwen2vl.hpp:131
↓ 2 callersFunctionmakeVisualRotaryPosEmb
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:132
↓ 2 callersFunctionmakeVisualRotaryPosEmbFull
mllm/models/qwen2vl/modeling_qwen2vl.hpp:95
↓ 2 callersFunctionmakeVisualRotaryPosEmbFull
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:96
↓ 2 callersFunctionmakeVisualRotaryPosEmbIds
mllm/models/qwen2vl/modeling_qwen2vl.hpp:29
↓ 2 callersFunctionmakeVisualRotaryPosEmbIds
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:30
↓ 2 callersFunctionmakeVisualRotarySinCos
mllm/models/qwen2vl/modeling_qwen2vl.hpp:109
↓ 2 callersFunctionmakeVisualRotarySinCos
mllm/models/qwen2vl/modeling_qwen2vl_traceable.hpp:110
↓ 2 callersFunctionmappingThis2NextPos
algorithms/lazy_vlm/models/qwen2_5vl/modeling_qwen2_5vl.hpp:27
↓ 2 callersFunctionmappingThis2NextPos
algorithms/lazy_vlm/models/qwen2_5vl/modeling_qwen2_5vl_fast.hpp:27
↓ 2 callersFunctionmappingThis2NextPos
algorithms/lazy_vlm/models/qwen2vl/modeling_qwen2vl.hpp:27
↓ 2 callersFunctionmat_mul
* NOTE: This only considered Linear operation with gguf quantizated weights, where weight is a 2D matrix * As Matmul Op is performed in float data ty
mllm/backends/cpu/kernels/common/ggml/matmul.cpp:41
↓ 2 callersMethodmax
(self, dim: int = -1, keep_dim: bool = False)
pymllm/ffi/__init__.py:196
↓ 2 callersMethodmax
mllm/core/Tensor.cpp:418
↓ 2 callersFunctionmemscpy
mllm/backends/qnn/QNNModel.cpp:352
↓ 2 callersMethodmin
(self, dim: int = -1, keep_dim: bool = False)
pymllm/ffi/__init__.py:193
↓ 2 callersFunctionmllmDataTypeToQnnDataType
mllm/backends/qnn/QNNUtils.cpp:303
↓ 2 callersFunctionmllm_cpu_has_neon
mllm/backends/cpu/kernels/arm/linear/ggml/gemm_aarch64.cpp:24
↓ 2 callersFunctionnextPowerOf2
mllm/backends/cpu/ops/STFTOp.cpp:44
↓ 2 callersMethodpad
Pad the image to target size with given RGB color. Semantics mirror PIL ImageOps.pad: resize to fit within target (keeping aspect ratio) then center t
mllm/preprocessor/visual/Image.cpp:156
↓ 2 callersMethodperfSummary
mllm/models/ARGeneration.cpp:268
↓ 2 callersFunctionpermute_fp16
mllm/backends/cpu/kernels/arm/permute.cpp:74
↓ 2 callersFunctionpermute_fp32
mllm/backends/cpu/kernels/arm/permute.cpp:19
↓ 2 callersMethodpreGetKVWriteLocation
mllm/nn/lmcache/StaticCache.cpp:157
↓ 2 callersMethodprefetch
mllm/engine/prefix_cache/Cache.cpp:35
↓ 2 callersMethodprepare
( self, quantize_config, tensor_dict: Dict, **kwargs )
pymllm/quantize/quantize_pass.py:20
↓ 2 callersFunctionprepare_kernel_args
mllm-kernel/mllm_kernel/cpu/csrc/add_constant.cpp:60
↓ 2 callersFunctionprint_stack_trace
mllm/mllm.hpp:313
↓ 2 callersMethodpromote
mllm/nn/lmcache/PrefixCache.cpp:20
↓ 2 callersFunctionpromoteMMAPTensor2AnonymousMemoryTensor
mllm/engine/io/CpuMemoryDiskDispatcher.cpp:29
↓ 2 callersFunctionpropagateQuantScale
mllm/backends/qnn/QNNUtils.cpp:615
↓ 2 callersMethodpull_file
Pull file from device
pymllm/utils/adb.py:265
↓ 2 callersMethodpurge
mllm/engine/prefix_cache/Cache.cpp:37
↓ 2 callersMethodpush
===----------------------------------------------------------------------===// RequestPool ===--------------------------------------------------------
mllm/engine/service/Service.cpp:16
↓ 2 callersFunctionquantize_row_q4_0
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q4.cpp:70
↓ 2 callersFunctionquick_gelu_fp32
mllm/backends/cpu/kernels/arm/gelu.cpp:82
↓ 2 callersFunctionreadWAV
mllm/preprocessor/audio/Audio.cpp:489
↓ 2 callersMethodrearrangeCache
mllm/backends/qnn/aot_rt/KVCacheManager.cpp:177
↓ 2 callersMethodrebind
mllm-kernel/include/mllm_kernel/tensor.hpp:338
↓ 2 callersFunctionreduce_max_fp16
mllm/backends/cpu/kernels/arm/reduce.cpp:51
↓ 2 callersFunctionreduce_max_fp32
mllm/backends/cpu/kernels/arm/reduce.cpp:24
↓ 2 callersFunctionreduce_max_int16
mllm/backends/cpu/kernels/arm/reduce.cpp:102
↓ 2 callersFunctionreduce_max_int32
mllm/backends/cpu/kernels/arm/reduce.cpp:126
↓ 2 callersFunctionreduce_max_int8
mllm/backends/cpu/kernels/arm/reduce.cpp:78
↓ 2 callersFunctionreduce_mean_fp16
mllm/backends/cpu/kernels/arm/reduce.cpp:145
↓ 2 callersFunctionreduce_mean_fp32
mllm/backends/cpu/kernels/arm/reduce.cpp:138
↓ 2 callersFunctionreduce_mean_int16
mllm/backends/cpu/kernels/arm/reduce.cpp:159
↓ 2 callersFunctionreduce_mean_int32
mllm/backends/cpu/kernels/arm/reduce.cpp:165
↓ 2 callersFunctionreduce_mean_int8
mllm/backends/cpu/kernels/arm/reduce.cpp:153
↓ 2 callersFunctionreduce_min_fp16
mllm/backends/cpu/kernels/arm/reduce.cpp:58
↓ 2 callersFunctionreduce_min_fp32
mllm/backends/cpu/kernels/arm/reduce.cpp:30
↓ 2 callersFunctionreduce_min_int16
mllm/backends/cpu/kernels/arm/reduce.cpp:108
↓ 2 callersFunctionreduce_min_int32
mllm/backends/cpu/kernels/arm/reduce.cpp:132
↓ 2 callersFunctionreduce_min_int8
mllm/backends/cpu/kernels/arm/reduce.cpp:84
↓ 2 callersFunctionreduce_sum_fp16
mllm/backends/cpu/kernels/arm/reduce.cpp:37
↓ 2 callersFunctionreduce_sum_fp32
mllm/backends/cpu/kernels/arm/reduce.cpp:12
↓ 2 callersFunctionreduce_sum_int16
mllm/backends/cpu/kernels/arm/reduce.cpp:90
↓ 2 callersFunctionreduce_sum_int32
mllm/backends/cpu/kernels/arm/reduce.cpp:114
← previousnext →701–800 of 5,971, ranked by callers